FinTechCryptoTask Prompt

Compliance Prompt for OFAC Sanctions Impact Analysis

This prompt takes a fresh OFAC sanctions update and maps it against your high-risk customer segments, surfacing every customer with more than 10% ownership overlap with a newly designated entity. It returns risk-tier reclassifications, prioritized P0/P1/P2 actions including transaction-freeze recommendations, and the audit-trail evidence each disposition decision requires. It is designed for sanctions compliance teams at fintech and crypto firms responding to same-day list changes.

How to use this prompt

1
Paste the prompt into Luna in your deepidv dashboard for live screening context, or into Claude, ChatGPT, or Gemini for a desk analysis.
2
Replace [Insert Date] with the OFAC update date and [Insert Region] with the sector or region you screen, then provide your high-risk segment data and ownership records from KYB filings or corporate registries.
3
Expect four sections back: affected customers with SDN references, risk-tier changes with effective dates, immediate actions, and audit-trail requirements.
4
Action P0 items immediately: freeze recommendations and regulator notifications come with stated rationale, so route them straight to your sanctions officer for sign-off.
5
Schedule the P1 and P2 re-verification work (KYC refresh, UBO re-discovery, enhanced due diligence) and anchor the disposition evidence where your audit trail lives.

The prompt

Luna, analyze the latest OFAC sanctions update from [Insert Date]. Map these changes against our high-risk customer segments in the [Insert Region] sector. Identify any entities with over 10% ownership overlap and suggest immediate re-verification steps.

For each impacted segment, return:

1. AFFECTED CUSTOMERS
- List of customer IDs with >10% ownership overlap with newly designated entities
- For each: counterparty name, jurisdiction, designation reason (SDN list category)
- Ownership stake percentage and source (corporate registry, prior KYB filing)

2. RISK CLASSIFICATION CHANGE
- Current risk tier vs recommended new tier
- Trigger event that justifies the change
- Effective date for the reclassification

3. IMMEDIATE ACTIONS
- Re-verification scope (KYC refresh, KYB UBO re-discovery, enhanced due diligence)
- Transaction freeze recommendation (if applicable) with rationale
- Notification requirements (regulator, internal compliance, counterparty)

4. AUDIT TRAIL REQUIREMENTS
- What evidence must be captured for the disposition decision
- Where the cryptographic receipt should be anchored

Be specific. Cite the OFAC SDN list entry, the EU Consolidated entry, or the UN sanctions reference where they apply. Do not hedge. For each customer, state whether they require P0 (immediate), P1 (within 24 hours), or P2 (within 7 days) action based on the input provided.

Test it in Claude or another LLM

This prompt is built for the Luna agent inside deepidv, where it maps a fresh OFAC sanctions update against your high-risk customer segments, flags every customer with more than 10% ownership overlap with a newly designated entity, and returns prioritized P0/P1/P2 re-verification steps. Here is how to dry-run the same workflow in any LLM first, using fake screening data, before you wire it to live deepidv customer records.

1
Paste the full prompt into Claude, ChatGPT, or Gemini and replace the opening "Luna," with a role instruction such as "Act as an OFAC sanctions compliance analyst." Also fill the bracketed fields: set [Insert Date] to a test date like 2026-06-01 and [Insert Region] to a sector such as "crypto remittance, MENA corridor."
2
Paste the synthetic sample data block below directly under the prompt so the model has fake customers, ownership stakes, and a fabricated SDN entry to reason over.
3
Good output for this prompt is four clearly separated sections: an affected-customers list that cites the (fake) SDN entry and states each ownership stake percentage, a risk-tier change with a trigger event and effective date, immediate actions including any transaction-freeze recommendation, and audit-trail evidence requirements. Every affected customer must carry a P0, P1, or P2 label rather than an undifferentiated list.
4
Iterate on the prompt wording until the priority labels and the SDN citation format come back clean and consistent across runs.
5
Once the output shape is right, run it live in the deepidv dashboard where Luna executes it against your real customer segments and current OFAC, EU Consolidated, and UN sanctions feeds, anchoring each disposition to your audit trail.

Synthetic sample data to paste alongside the prompt

Fake test data, safe to share with any LLM. Swap in your own once the output looks right.

OFAC SDN UPDATE (TEST, 2026-06-01): Added entity ACME-HOLDINGS-TEST-LLC, jurisdiction Testistan, SDN program [TEST-SDGT].
Customer CUST-TEST-001 (Jane Testcase) ,  owns 14% of ACME-HOLDINGS-TEST-LLC per KYB filing FAKE-REG-2024-0000; current risk tier: Standard.
Customer CUST-TEST-002 (Sample Trading Co) ,  owns 8% of ACME-HOLDINGS-TEST-LLC per corporate registry stub; current risk tier: Enhanced.
Customer CUST-TEST-003 (Placeholder Ventures) ,  owns 51% of unrelated entity DEMO-ENTITY-999 (not listed); current risk tier: Standard.

Pairs with on deepidv

Sources & further reading

FAQ

What is the OFAC 50 percent rule and why does this prompt use a 10% ownership threshold?

OFAC's 50 Percent Rule blocks any entity owned 50% or more, individually or in aggregate, by sanctioned persons, even if that entity is not itself listed. Many compliance teams screen at a lower threshold like 10% as an early-warning buffer, because aggregated stakes and indirect ownership chains can cross 50% quickly and the cost of a missed blocking obligation is severe.

How fast do I need to act when OFAC adds a customer's owner to the SDN list?

Blocking obligations take effect immediately upon designation, so any property or transactions involving a newly blocked entity must be frozen the same day and reported to OFAC within 10 business days. That is why this prompt forces a P0/P1/P2 priority on every affected customer instead of returning an undifferentiated list.

Run it with live verification data

These prompts work in any LLM. Inside the deepidv dashboard, Luna, Arbiter, and Arc run them against your real sessions, screening lists, and audit trails.

Book a Demo